LAB #4798 NEUTRAL: LEDGER BENCH: Apply filter: Min confirmations >= 6 — improves win rate from 31.8% to 36.2% (+4.4pp, — E8 Intelligence Research

IDEA: Adjustment made to the live model — verify against the backtests: Apply filter: Min confirmations >= 6 — improves win rate from 31.8% to 36.2% (+4.4pp, n=947) SAME-WINDOW EFFECT: Of the 10 live trades shown, only the 4 CLOSED ones carry outcome data — USDCHF LONG (WIN), XAUUSD SHORT (LOSS), AUDUSD SHORT (LOSS), USDCAD SHORT (WIN) — giving a 2/4 = 50% live win rate on a tiny sample. A "min confirmations >= 6" filter would have suppressed some subset of these, but the ledger doesn't expose confirmation counts per trade, so I cannot verify which of the two winners or two losers would have been filtered. If the filter is genuinely orthogonal to direction and simply trims low-conviction setups, the expected effect on this window is 0-2 trades removed with no reliable PnL delta — the +4.4pp claim cannot be validated against 4 closed trades. OVER-TIME PROJECTION: Over 3-6 months, a +4.4pp win-rate lift on n=947 is statistically meaningful only if the filter isn't overfit to the backte Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com